BUG stoch_vol model returns two arguments now.

Author: twiecki

pyfolio/plotting.py

@@ -419,7 +419,7 @@ def plot_drawdown_periods(returns, top=10, ax=None, **kwargs):
 
     ax.set_title('Top %i Drawdown Periods' % top)
     ax.set_ylabel('Cumulative returns')
-    ax.legend(['Portfolio'], 'upper left')
+    ax.legend(['Portfolio'], loc='upper left')
     ax.set_xlabel('')
     return ax
 

pyfolio/tears.py

@@ -838,14 +838,14 @@ def create_bayesian_tear_sheet(returns, benchmark_rets=None,
 
     if stoch_vol:
         # run stochastic volatility model
+        returns_cutoff = 400
         print(
             "\nRunning stochastic volatility model on "
-            "most recent 400 days of returns."
+            "most recent {} days of returns.".format(returns_cutoff)
         )
-        returns_cutoff = 400
         if df_train.size > returns_cutoff:
             df_train_truncated = df_train[-returns_cutoff:]
-        trace_stoch_vol = bayesian.model_stoch_vol(df_train_truncated)
+        _, trace_stoch_vol = bayesian.model_stoch_vol(df_train_truncated)
         previous_time = timer(
             "running stochastic volatility model", previous_time)